Pranav Piyush

The martech industry is broken, and I’m devoting the next 15 years of my life to fix it.

After a decade leading growth and marketing teams at places like Dropbox, Pilot, Magento and BILL, I realized there was no measurement tool on the market that was statistically valid, simple and holistic. 

So I went and talked to every CMO and VP of marketing in my network. Their number one complaint? Attribution. 

And this is why we built Paramark.

Because I believe that measuring your marketing efforts is not that hard. 

Through the power of MMM, experimentation and data science, we’ve created the holy grail of measurement tools. So marketers like you can confidently understand the impact of your marketing investments – and communicate them to your CFO and CEO.

Learn our method at Paramark.com.

Co-Founder & CEO

Before starting Paramark, I spent the last decade leading growth and marketing teams at fast-growing companies like Dropbox, Pilot, Magento, and BILL.

During that period, we evaluated hundreds of martech tools and, for the most part, were sorely disappointed. Here's why. We found few solutions that could handle the needs of fast-growing marketing teams. They were either incomplete, hard to use, or claiming things that were just flat-out wrong!

This trend is especially true for marketing analytics. Synthesizing and analyzing marketing data to understand effectiveness should be a solved problem. After speaking to hundreds of marketing leaders at some of the fastest growing companies, I was honestly surprised that the truth is far from it. Understanding and communicating the return on marketing investment continues to be one of the top challenges for marketers.

That's why we decided to build Paramark. Our goal is to help fast-growing marketing teams measure and forecast the impact of their investments, unlocking sustainable growth.

10 marketing predictions for 2026 that I'm willing to bet money on. Trigger warning... this ain't for the faint of heart.

1. The volume of em-dash hating posts will not reduce this year

2. Gullible marketers will keep being sold advanced attribution and ai software to solve their growth problems

3. Most marketers will not speak to any real customers or do any real research for weeks (or even months) at a time

4. Also, most marketers will STILL not study the fundamentals of psychology or economics to improve their work

5. Most marketers will continue NOT influencing pricing, product, or business model and yet be given larger and larger revenue goals

6. Many will try using AI for content and then give up when it produces sloppy output and then whine about how AI is no good

7. Lots of gurus on LinkedIn and X will claim they've found the trick for 10x-ing growth in 2026 with some new advertising or growth strategy

8. Brand marketers who don't embrace measurement will struggle to make an impact

9. Performance marketers who don't embrace long-term thinking and "brand building" will continue to hit diminishing returns

10. CMOs who don't spend time educating the CFO and CEO on how marketing actually works, while also driving clear evidence-based growth, will lose their jobs

CEO @ Paramark | Marketing measurement that CFOs & CMOs trust

Marketers deploy $1 trillion/year in capital. Very little of it is analyzed correctly.

What's worse? There's an additional $500B spent on ops, staff, research, and analytics... and it's still a mess.

We’re fixing that — and 2026 is when it accelerates.

If you’ve ever wanted to help build the operating system for modern marketing, this is the year.

In 2026, we’re making three big investments:

1. Product. Expanding our platform to radically shorten the path from idea to launch, with finance-grade measurement our customers already trust.

2. Go-to-market. Scaling sales with senior leadership, new hires in SF and NY, and a dedicated events & partnerships motion.

3. Solution engineering. Growing the team that owns the full data stack — from validation and modeling to automation — turning complex marketing data into decisions people can act on.

If this excites you, check out our careers page... or help us spread the word. 🙏
https://lnkd.in/giVpNhiQ

Founder @ 42 - Paid Media, Creative, SEO & RevOps for B2B

Help B2B companies scale @ 42 Agency
Helping inter-generational families connect with learning @ Little Robin Hood
Build better exclusions for online ads @ exclusion sites 
Publisher and writer  on B2B marketing insights @ 42slash.com 
Various Projects at rext.in

Kamil Rextin

I have been hacking away at Claude Code a lot over the holidays and it's official: I'll be hosting a workshop on Claude Code with Exit Five  Thank you so much to Matthew Carnevale  for making it happen.

If you're curious about how Claude Code can help you as a marketer, doing anything from data analysis to generating content, building custom agents, would love to see you there. Let me know if you have any questions you want me to address or use cases that are most interesting / curious about!

Attribution said paid social drove 257 leads for a B2B SaaS client.

The real number was closer to 2,137.

That's an 8.3x blind spot.

Here's how we found it. We matched companies that had any LinkedIn ad exposure against their CRM data—not just clicked. Not just converted. Any exposure.

Companies with LinkedIn exposure had a 6.72% opportunity rate. Companies without? 0.83%.

That's an 8.1x lift. Statistically significant.

Your crm model is lying to you. Not because it's broken. Because it was never designed to measure what paidsocial actually does.

paidsocial doesn't convert people in-session. It makes your brand familiar. It warms up accounts. It does the work that later appears as "direct" or "organic" or the sales team taking credit.

If you're measuring paidsocial the same way you measure search, you're measuring it wrong.

The question isn't whether LinkedIn is working. It's how much credit you're giving to channels that are just harvesting what LinkedIn planted.

I don't want to explain my content system every time I use it.

So I saved it as a Claude Code agent.

It knows the 5 themes. It knows the tease-publish-promo rhythm. It knows each essay spawns 6-9 LinkedIn posts and 1-2 carousels. It knows to hit the ClickUp API with the right tags, due dates, and assignments.

Next time I need content? I invoke the agent. It doesn't ask clarifying questions about structure. It just builds.

This is the shift most people miss about AI. Everyone's focused on "prompting better." That's a trap. You're doing the same work over and over.

Build systems, not prompts.

We spent the time once to encode our entire content engine - 10 months, 20 essays, 200+ derivatives - into something reusable. Now it's a tool, not a task.

The first time takes effort. The second time takes seconds.

What if your best processes were preserved and reusable instead of stuck in someone's head?

No alternative text description for this image

I built a paid media co-pilot this weekend with Claude Code. Here's what actually matters.
Most of what it does is table stakes - unified dashboard across LinkedIn/Meta/Google, performance tracking, ad variation generation, approval workflows.

But there's one piece that fundamentally changes how I think about audience building:
Natural language audience targeting for LinkedIn.
Here's the problem we've all experienced:
You know exactly who you want to target. "Marketing leaders at mid-market B2B SaaS companies focused on demand generation."
But translating that mental model into LinkedIn's targeting interface means:

Clicking through job function taxonomies
Searching job titles one by one
Guessing which seniorities to include
Hoping you didn't miss relevant segments
No way to validate you captured your actual ICP

The new workflow:

Describe your audience in plain language
The tool analyzes your website + description
Queries a database of LinkedIn's complete targeting taxonomy (job functions, titles, seniorities, skills)
Matches and ranks relevance (highly relevant, partial matches, not relevant)
Shows you the API targeting script it will generate
Saves the audience for reuse
Pushes it directly to LinkedIn as a draft campaign

Why this matters:
When you stop treating ad platforms as the primary interface and start treating APIs as the interface, your workflow changes from searching to describing.

You're no longer constrained by platform UX. You're expressing intent and letting systems translate that into platform-specific parameters.
This is the same shift that happened with SQL → natural language for databases.
The platform interface becomes a validation layer, not the creation layer.
What I learned building this:
LinkedIn's API doesn't return audience penetration (annoying).
Demographics require a separate API that returns URNs but not company names (very annoying).
But the targeting taxonomy is comprehensive if you structure it properly.
And Claude Code made building this dramatically faster than traditional development would have been.

Shout out to the 42 Agency team, especially Ma Cherrie Dee Gonzales and Alejandra Céspedes, for creating the custom box and designing the it

Vanessa Griffiths & team  awesome partners to work with on this .

If you want an awesome giftbox, send me a DM (if you need RevOps or Demand Gen support) & I will ship you one personally.

Fractional CMO for Seed to Series B startups | Founding marketing leader at Float & CrowdRiff 📈 | Builder of standout startup brands

I already thanked them privately but nice touch from 42 Agency on their client gifting program. You don't have to be a big company to do things well. (And of course the bigger thanks is for their help untangling some thorny MOPS problems for me!). Well done Kamil Rextin Alejandra Céspedes.

NEW ESSAY: Paid Media Isn't Your Problem (Coming January 20th)

Last quarter, a B2B SaaS company came to us: "Our paid media isn't working."

We pulled their CRM. 47% of leads were sitting in "open" status. Never worked.

Their paid wasn't broken. Their system was.

This essay breaks down the 3 problems disguised as paid media problems—and introduces Compound GT

We just published a new essay on our Substack: 

Why most paid media problems are not actually paid media problems but they're more measurement or GTM stack problems. 


Check out the link below. Let me know if it interesting or if you have any feedback

taplio

The Battle

Pranav Piyush

Kamil Rextin

Followers

Posts

Engagements

Likes

Performance ScorePranav Piyush's LinkedIn Performance

Performance ScoreKamil Rextin's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Pranav Piyush

Kamil Rextin

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Pranav Piyush's LinkedIn Performance

Performance ScoreKamil Rextin's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScorePranav Piyush's LinkedIn Performance